摘要
为了保证煤矿安全开采,并提高煤矿瓦斯涌出量的预测精度,提出了改进思维进化算法优化BP神经网络的模型预测新方法。在思维进化算法中加入精英反向学习策略增加算法的全局搜索能力,在趋同操作中引入粒子群算法避免重复搜索,以此实现对BP神经网络的初始权值和阈值的全局寻优,并通过矿井监测到的各项历史数据进行验证。结果表明:与BP神经网络模型和MEA-BP神经网络模型相比较,该模型的预测精度更高,泛化能力更强。该模型的平均相对变动值为0.00116,平均相对误差为0.81%,均方根误差为0.0576,有效提高了对瓦斯涌出量的预测精度,提升了煤矿安全生产技术。
In order to improve coal mine safety and mining technology and improve the prediction accuracy of coal mine gas emission,it proposed A new mothod of model prediction was proposed based on BP neural network by improved mink evolutionary algorithm(MEA).The elite opposition-based learning was added to MEA to enhance the global search ability.The particle swarm optiomization algorithm was introduead to avoid repeated searches.In this way,it realized the global optimization of the weights and threshods of BP neural network and verified through various historicla datas monitored by the mine.The results showed that this model had higher pediction accuracy and stronger generalization ability compared with BP neural network model and MEA-BP neural network model.The average relative variance of this model was 0.0016,average relative error was 0.81%,and root mea square error was 0.0576,which effectively improved the prediction accuracy of gas emission and improved the coal mine safety production technology.
作者
赵焕平
ZHAO Huan-ping(School of Computer and Software,Nanyang Institute of Technology,Nanyang 473004,China)
出处
《南阳理工学院学报》
2023年第4期35-39,共5页
Journal of Nanyang Institute of Technology
基金
河南省科技攻关项目(142102210554)。
关键词
瓦斯涌出量
思维进化算法
精英反向学习
粒子群算法
BP神经网络
gas emission
mind evolutionary algorithm
elite opposition-based learning
particle swarm optimization algorithm
BP neural network
